Targeted Nonlinear Adversarial Perturbations in Images and Videos

Abstract

We introduce a method for learning adversarial perturbations targeted to individual images or videos. The learned perturbations are found to be sparse while at the same time containing a high level of feature detail. Thus, the extracted perturbations allow a form of object or action recognition and provide insights into what features the studied deep neural network models consider important when reaching their classification decisions. From an adversarial point of view, the sparse perturbations successfully confused the models into misclassifying, although the perturbed samples still belonged to the same original class by visual examination. This is discussed in terms of a prospective data augmentation scheme. The sparse yet high-quality perturbations may also be leveraged for image or video compression.

Code and data available at: https://github.com/roberto1648/adversarial-perturbations-on-images-and-videos

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Explaining and Harnessing Adversarial Examples
• The Kinetics Human Action Video Dataset
• Striving for Simplicity: The All Convolutional Net
• Delving into Transferable Adversarial Examples and Black-box Attacks
• Data Augmentation Generative Adversarial Networks
• Object Detectors Emerge in Deep Scene CNNs
• A Closer Look at Memorization in Deep Networks
• Deep Image: Scaling up Image Recognition
• Adversarial Transformation Networks: Learning to Generate Adversarial Examples
• Sparse Adversarial Perturbations for Videos

Cited by in corpus (3)

• Black-box Adversarial Attacks on Video Recognition Models
• Over-the-Air Adversarial Flickering Attacks against Video Recognition Networks
• Regularized adversarial examples for model interpretability